• Hey Trainers! Be sure to check out Corsola Beach, our newest section on the forums, in partnership with our friends at Corsola Cove! At the Beach, you can discuss the competitive side of the games, post your favorite Pokemon memes, and connect with other Pokemon creators!
  • Due to the recent changes with Twitter's API, it is no longer possible for Bulbagarden forum users to login via their Twitter account. If you signed up to Bulbagarden via Twitter and do not have another way to login, please contact us here with your Twitter username so that we can get you sorted.

Mafia Philosophy

Status
Not open for further replies.
You get better at riding your bike, by riding your bike. You don't necessarily improve by having people push your bike from behind forever. Eventually, you need to just pull the trigger and take the risk of actually pedaling on your own. The more you do it, the better you get at it. Nobody starts off fantastic, and most people start off pretty terrible and awkward at it.

I wouldn't suggest that you avoid playing mafia, I just want you to get better so that you can get more out of playing mafia.

Go back and read the screw up that we call XY Mafia... in which the scum win was entirely my fault (If I had followed @Hellcrow's advice, we would have won but I didn't trust HC).... That game taught me that I'm better off in the shadows than on the front lines.
 
Alright, let’s go through this again:

Why was No Lynching Day 1 (or in general) frowned upon Bulbagarden for many years?

It first started in a small game (8 or 9 people) with Phoenicks in it as mafia. Many of you may not know him, he’s an old player who rarely shows up these days, but he is very good. After lots of No Lynches in a game he was scum, he sweeped it. think of it as a very old version of MLP

Sometime around June 2014, Gundam SEED mafia was hosted by FinalArcadia. 18 people, 13 town, 4 scum, 1 indep. Lots of No Lynches in it, including one I caused with my role to avoid a mislynch on myself d1 (which may have been a misplay now that I think of it). HumanDawn was scum in it. Clean sweep. Some people who still thought No Lynches were good changed their mind after that game and the one I’ll mention below. HumanDawn has always been anti-No Lynch, in that game he kinda toned it down to skate around the game and clean-sweep it with his team. think of it as a 2014 version of MLP that had more players

A similar game that was very inactive with many No Lynches and HumanDawn scum was hosted in July. Collapse, hosted by Beck…

And now, we got MLP…

Tl;dr NLs harm town far more than they benefit it.

Especially in some games like 9 players 2 mafia, they are more harmful because they cause a LYLO become a 1-day-earlier-MILO…

About vanilla townies:

In Mafia Universe, the most common setup is the 9-man open vanilla.

1 cop, 6 vanilla town, 2 vanilla mafia

YOU ARE NOT ALLOWED TO NO LYNCH. If votes are tied a random person with most votes is lynched. This ensures that MILO can never happen, and if no town sweep happens, LYLO is guaranteed to happen.

13-man vanilla (1 cop, 9 vanilla town, 3 vanilla mafia) is also common enough, to the point it was selected as the setup under which the mafia championship was played this year. This means that it is an undeniably balanced setup.

I’ve even seen 17-man and 21-man setups which are similar (m mafia, 3m vanilla town, 1 cop).

There are other variations with many vanillas, including one of the hardest open setups ever for town, the Mountainous one, in which ALL players are vanilla (13p, 11 town, 2 mafia)

Finally, about activity:

I’ve been keeping statistics over the games I’ve played and their activity.

There are three ways to see how active a game is:

P: Posts. How many posts a game has. Very easy to check since you can just see how many replies the game thread has. Indicative of how big a game is, but not really of how active it is, because you can’t really compare a 9p with a 20p game like this.

P/P: Posts per Player. Better measure of activity. Indicates how many posts did the average player make in the game. Still not completely indicative enough because in a small game you don’t have much to talk about, there only a few players you can read and interact with. So…

P/P2 Posts per Player Squared: A different measure of activity, also good but has its flaws. Indicates how many posts did each player invest to interact with each of their fellow players. Has a lot of error because the hosts’ posts are counted and a player doesn’t interact in-thread with themselves.

I’ll give you a table of these things for our recent games in a while.
 
Go back and read the screw up that we call XY Mafia... in which the scum win was entirely my fault (If I had followed @Hellcrow's advice, we would have won but I didn't trust HC).... That game taught me that I'm better off in the shadows than on the front lines.

Not exactly... the first screw up happened in Day 3 in which I was still alive. You didn't trust me enough.

100% of the info Hellcrow had was given to him by me, after I've explained lots of times why he was town. You still didn't trust him.

The entire faction paid the cost of that, as well as many modkills on townies (including one on the cop).

Nonetheless, this should've taught you how to become a better leader, not push you to the sidelines.

It's been 5 times scum!HumanDawn has outplayed me, I should've been afraid to speak at this point, let alone lead if need be. That wouldn't really help me become better...

I mean, if @Zexy votes me day 1, is he joke voting? Or is he mafia, trying to off an important town role?

Just using Zexy as an example
C:\Users\Dinos\AppData\Local\Temp\msohtmlclip1\01\clip_image001.png

A bad example :)

In the past, I tried to joke vote strategically... then I realised you can't really use joke votes strategically, plus they only confuse things leading to even more d1 mislynches.

At this point, I hate that stage with a burning passion. I would never just joke vote. This doesn't mean I have to be mafia trying to off you, though, what if I am town that honestly suspects you? :p

… the problem Bulba has is that scumhunting was based too much in outside chat and role interactions, a thing called “night chess” in other sites. It’s like you have an asset of night actions and try to make the best “chess” move to win.

As I said, I’ll show you activity numbers like P P/P and P/P2 of vanilla games from other places. Day 1 is extremely important there, lots of posts happen and even more scumhunting happens, with 90 or even outright 100% vanillas in Mountainous. Even huge games like 57p or 85p have tons of vanillas, this doesn’t make things harder for town.
 
Game, P, P/P, P/P2
Whoniverse 838 39.9 1.9
Murder She Wrote 217 27.1 3.3
Hogwarts 461 28.8 1.8
Christmas 481 32 2.1
FE Protagonist 645 32.25 1.6
MLP 211 23.4 2.6

4 vanilla 9ers from MU (TURBO, as in, the entire game was played in under an hour with 18 min days/6 min nights)
389 43.2 4.8
341 37.8 4.2
476 52.8 5.9
376 41.7 4.6

And 3 other games from MU. 85p 57p and an 27p I wanna show you because of its rare numbers
85p 15564 183.1 2.15
57p 15570 273.3 4.79
27p 7806 289.1 10.7

A few things to note:

The biggest game in Bulba, P wise, was Bulba II mafia by Mijz with 1640 posts. Most big enough MU games easily surpass this, with the biggest MU game P wise being the 57p with 15570 posts!

The most active game in Bulba, P/P wise, was RM, with a P/P of 73.5 Still, most big enough MU games surpass this, and 9er turbos which last less than an hour surpass medium-sized Bulba games that last days. I won’t even bother comparing them to 9ers in Bulba like Murder She Wrote or MLP…

MLP's P/P is very bad. Hogwarts and Murder She Wrote also had bad P/P. The former was won by mafia, the latter was won by town because the setup was solvable and a town leader rose up to break it open using lots of private communications. Still very inactive. Other games won by mafia have P/P or 11, 16 or 17 (ARC-V, Collapse, VGM).

Which game out of all the ones I’ve shown you would you call the most active? The 85p was considered a beast at the time it was hosted, then the mafia won it, and the townies realized they had to put more effort in it. When the 57p game hit, they took it to overdrive, achieving 6 more posts with 30 less players. The game was still won by mafia in the end but was much more closer, ending in a 3 town-2 mafia LYLO.

The third game is easily the most active one I’ve seen, even though it only has half the posts of the 85er or the 57er. The P/P is just 10 more than the 85ers, but the P/P2 is more than DOUBLE! And this game, despite its activity (or maybe because of it) was a very good town sweep in which the last Mafiosi conceded (had they not it would be even bigger!). This shows that neither P/P nor P/P2 are completely accurate in calculating activity, but it also shows that LOTS of activity is needed for town to win a game, and that if town plays correctly, many vanillas aren’t really a problem.
 
@Zexy What calculation did you use to calculate P/P^2?

Also speaking of the Videogameverse thing on MU; I'll switch to player if you play :) shut up caps you just signed up for code geass on pc and two games in here you're going to die
You calculate P/P, have it hold an Up-Grade, then trade it. And if you trade it again while holding a Dubious Disc, it becomes P/P-Z :p
Seriously, it's what it says on the tin. Posts/amount of players squared. For a regular 9er it is Posts/81.

I won't play in it because of finals and many sign-ups here. you should indeed shut up caps
 
You calculate P/P, have it hold an Up-Grade, then trade it. And if you trade it again while holding a Dubious Disc, it becomes P/P-Z :p
Seriously, it's what it says on the tin. Posts/amount of players squared. For a regular 9er it is Posts/81.

I won't play in it because of finals and many sign-ups here. you should indeed shut up caps

...wow I'm dense

Mmkay. I signed up as a sub. salt salt salt salt salt salt
 
You calculate P/P, have it hold an Up-Grade, then trade it. And if you trade it again while holding a Dubious Disc, it becomes P/P-Z :p
Seriously, it's what it says on the tin. Posts/amount of players squared. For a regular 9er it is Posts/81.

I won't play in it because of finals and many sign-ups here. you should indeed shut up caps

As if you know when it's even happening
 
I never know when anything is happening. I just kinda sit back and stuff happens.

Also, getting back to the D1 Lynch topic, I used to be very opposed to a D1 Lynch, but that was mostly because I always ended up being said D1 Lynch, regardless of whether I was Town or Mafia, and always for ridiculous or silly reasons.
(Source: Protagonist Mafia and Video Game Mafia)

However, I have recently come to acknowledge that D1 Lynches, if done right, can provide some useful information. That said, even a D1 No Lynch can provide plenty of information, assuming a fairly active phase. (Source: RMafia)
Anything that possibly gives Town some sort of information regarding setup can be very useful early on.
However, a D1 Lynch can also be dangerous if done completely at random. Roles like Bomb, Beloved, and/or any other negative utility role can be triggered and give the mafia a very early lead.
 
However, a D1 Lynch can also be dangerous if done completely at random. Roles like Bomb, Beloved, and/or any other negative utility role can be triggered and give the mafia a very early lead.
Or the town a very early lead, depending on who the bomb hits and if the Beloved skips the next night phase or day phase.
 
Beloved usually skips the next Day Phase, which means Town would lose 3 players right off the bat, assuming mafia NK works every night.
Bomb could blow up in a scum's face, but if the Bomb is the lynch target, it can just as easily blow up in a Townie's face which, again, means three potential Town deaths before the second Day Phase.
 
Not exactly... the first screw up happened in Day 3 in which I was still alive. You didn't trust me enough.

100% of the info Hellcrow had was given to him by me, after I've explained lots of times why he was town. You still didn't trust him.

The entire faction paid the cost of that, as well as many modkills on townies (including one on the cop).

Pretty much my point :) AussieEevee breaks things.

As for D1 lynching... Honestly, I wouldn't be so much in support of day 1 no lynch if joke voting wasn't a thing...

Do other sites suffer the joke voting problem?
 
Pretty much my point :) AussieEevee breaks things.

As for D1 lynching... Honestly, I wouldn't be so much in support of day 1 no lynch if joke voting wasn't a thing...

Do other sites suffer the joke voting problem?

Most sites I have played on run days for 72 hours (or more). This allows everybody to appear, cast their stupid RVS fuckvote, claim Miller or w/e if they need to and actually begin coming up with some [albeit strawgraspy] casing and association building. 48 hour phases here struggle to achieve that, from what I've been seeing, because by the time phase enters its final quarter, only 2/3 of the players have even appeared to post once or twice. If things were more active, joke voting wouldn't be a problem because there'd be more to react to which would end that stage of D1 much faster. That's why I'm for hosting games with 72 hour day phases.


In my short time here, I've seen games plummet in activity faster than a cannonball through a thin ice sheet. This means to me, that players who don't like this need to do stuff to try to end it and engage the inactive players before they settle into a mindset of "i'll just wait until something happens". Just Waiting can be the difference between lynching the SK on D1, or having the SK and Mafia hit the Cop & Doc on N1. It's not always going to happen, but in a general setup, there's a 20-30% chance that your first lynch, even blind, will hit scum and benefit town. Worst case scenario, you're looking at a 70-80% chance to hit town, of which the actuality of their role being important and vital to town success is lower. For instance, you're less likely to blind-lynch the Town Cop than you are to lynch the Mafia [Anything].


Joke Voting won't stop, but putting a snappy end to it is the responsibility of town. Scum is going to come out on top every single time jokevoting goes nowhere. Even if scum gets lynched, there are almost zero leads left to catch other scum on for D2 followups. Every non-lynch on scum is a benefit for them and makes town's job harder to recover. Info for scum is way more helpful for scum than info is for town, generally, since basically any info one scum acquires benefits an entire team, while say, town Tracker isn't sharing their info with the game unless they absolutely need to. Statistically speaking, the more posts you have, the more reads will exist. It's not hard to look at it as a numbers game. If everybody posts 3x per phase, or 30x per phase, you're more likely to have a better understanding of the players' and their relationships with others in the latter.
 
MU's not like Bulba. You can predict when a game is going to start very well.

AussieEevee, my point was that, despite the mess-up of that game, you should not be afraid of future ones and move on, in order to get better.

MU handles D1 completely differently from here; many posts that let you read everyone well enough and actually scumhunt from that early and decide for a lynch. There are not many inactives, and being inactive will guarantee your death sooner than later.

I don’t think Christmas was won by joke voting, more like Elie breaking the setup open…

Great analysis from Elie.
 
I said that because the joke voting early on helped confirm Elieson as town, allowing him to lead and give us the game. It was more of a joke than anything else.
 
I said that because the joke voting early on helped confirm Elieson as town, allowing him to lead and give us the game. It was more of a joke than anything else.
Not exactly. Elie and ME were pretty much consistently breadcrumbing since the beginning. It was disguised as jokes but served a very important purpose right off.
You can say the kill on ME was partly joke though.
 
Status
Not open for further replies.
Back
Top Bottom